The National Cancer Informatics Program (NCIP) has identified the critical need in the NCI-supported research community for wider access to large-scale data repositories, as well as the computational infrastructure necessary to facilitate integrative cancer research. It is essential that the computational capacity and the data repositories be co-located to minimize the heavy burdens associated with large-scale data transfers and the unnecessary mirroring of large-scale data sets.